AU782995B2 - Arginine mimetics as factor Xa inhibitors - Google Patents

Abstract

The invention relates generally to a novel type of arginine mimetics which are inhibitors of factor X<SUB>a</SUB>; to pharmaceutical compositions which comprise these mimetics; and to the use of these arginine mimetics for producing compositions for antithrombotic therapy.

Description

t WO 01/58859 PCT/EP01/014 2 3 Arginine mimetics as factor Xa inhibitors Description The invention relates generally to a novel type of arginine mimetics, which are inhibitors of factor Xa; to pharmaceutical compositions which comprise these mimetics; and to the use of these arginine mimetics for producing medicines for antithrombotic therapy.

Proteins, such as thrombin, which are involved in the blood coagulation cascade have for many years now been potential targets in the treatment of vascular diseases, with the aim of inhibiting of them and thereby avoiding thrombotic vascular occlusions or reopening thrombotically occluded blood vessels. The use of conventional anticoagulants which contain thrombin inhibitors is problematical since they increase the probability of bleeding complications Phillipides and J. Loscalzo (1996), Coronary Artery Dis., 7, 497-507). Furthermore, directly inhibiting thrombin does not interrupt the production of thrombin from prothrombin. In this case, therefore, it is necessary to supply relatively high doses of inhibitor continuously in order to maintain an antithrombotic effect in vivo.

Consequently, in the search for novel antithrombotic medicines, inhibition of the blood coagulation factor Xa became a main target for developing active compounds. Factor Xa is a trypsin-like serine protease which converts the zymogen prothrombin into its active form thrombin. By inhibiting factor Xa, therefore, it is possible to prevent thrombin being formed while a level of thrombin activity which is required for primary hemostasis is maintained Al-Obeidy and J.A. Ostrem (1998), Drug Discovery Today, 3, 223-231).

t-i -4 WO 01/58859 2 PCT/EP01/01 4 2 3 A large number of factor Xa inhibitors are nowadays known. The development of a number of these factor Xa inhibitors is based on conserving the structural motif Gly-Arg (with Gly being the P2 radical and Arg being the Pl radical, i.e. Gly binds in the S2 pocket and Arg binds in the S1 pocket of the factor Xa protein) at the site at which the prothrombin is cleaved by factor Xa.

In this connection, a large number of syntheses, in which a phenylalanine radical which is substituted on its phenyl ring by a basic amidino group is used as a mimetic for the Arg radical, have been described for factor Xa inhibitors StUrzebecher et al., (1989), Thromb. Res. 54, 245-252). It has been found that the factor Xa inhibitors in this series which have thus far been most effective are derivatives of 3'amidinophenylalanine.

Taking the guiding structure Na-tosylglycyl-D,L-3amidinophenylalanine alkyl ester (Compound 1 in Fig. 1; J. Sturzebecher et al., see above) as a starting point, a large number of peptidic bisbenzamidine compounds have been developed. The most powerful factor Xa inhibitor (Ki 0.5 iM) from this series is Na-4amidinobenzenesulfonylglycyl-D,L-4-amidinophenylalanine ethyl ester (Compound 2 in Fig. 1; B. Gabriel et al., (1998), J. Med. Chem. 41, 4240-4250), which binds "inversely" to factor Xa: its 4'-amidinobenzenesulfonyl group lies in the S1 pocket of factor Xa, -while the remainder of the molecule, together with the glycyl spacer, projects into the hydrophobic S3/S4 binding sites, thereby making possible additional interactions with the electronegative cavity, which is formed by the carbonyl oxygens of Lys 96, Thr. 98 and Glu 97 and its carboxylate group, behind the hydrophobic S3/S4 region.

According to the prior art, the group which is linked N-terminally to an amidinophenylalanine radical or to another arginine mimetic constitutes the P3/P4 radical of the potential factor Xa inhibitor, with the N- P.kOPER\M a2005U60164 amd pc doc-296~05 -3terminally linked group preferably being bonded to the arginine mimetic by way of a glycine spacer and a sulfonamide group.

Since then, nonpeptide bisbenzamidine compounds which possess markedly improved inhibitor properties (Ki 34 nM), and which are characterized by a shorter distance between the two aromatic groups, have been obtained Maduskuie et al., (1998), J. Med. Chem. 41, 53-62). On the basis of modeling analyses, it is assumed that an inhibitor 3 of this nature (see Figure 1) extends, by means of the m-benzamidine group, into the S1 pocket, and interacts in this pocket with the Asp 189 radical, and, by means of the p-benzamidine group, into the S4 aryl-binding pocket, where it enters into cation-n interactions and hydrophobic interactions with the surrounding radicals Phe 174, Tyr 99 and Trp 215.

In one aspect the present invention provides a factor Xa inhibitor which comprises an arginine mimetic which 20 possesses a N-terminal radical and a C-terminal radical, with the conformation of the inhibitor enabling intermolecular interactions to take place between the Cterminal radical and S3/S4 pocket of the factor Xa protein.

A binding mode of this nature, in which the C-terminal radical extends into the S3/S4 binding pocket of the factor SXa protein is particularly advantageous since, in the case of o the factor Xa inhibitor according to the invention, the Cterminal radical, in addition to the N-terminal radical, of the arginine mimetic also enters into intermolecular S 30 interactions with the factor Xa protein. Consequently, the factor Xa inhibitor according to the invention can be optimized both at the N-terminal radical and at the Cterminal radical, which means that it is possible to provide P:OPER\Vl4\205\25601 64 md spe doc-290605 -4inhibitor strengths which are markedly superior to those of the prior art.

A binding mode of this nature is surprising since all the arginine mimetic-based factor Xa inhibitors which have thus far been disclosed in the prior art bind in a substrate-like manner. In the substrate-like binding, the arginine mimetic binds in the S1 pocket while the radical which is linked Nterminally to the arginine mimetic by way of a potential P2 radical, such as a Gly spacer or the like, extends into the S3 or S4 pocket, respectively. In the case of the factor Xa inhibitor according to the invention, the arginine mimetic likewise binds in the S1 pocket but, in contrast to the factor Xa inhibitors known from the prior art, the radical which is bound C-terminally to the arginine mimetic, and not the radical which is bound N-terminally, extends into the S3 or S4 pocket of the factor Xa protein, respectively.

e eeo oee•.

I c WO 01/58859 5 PCT/EP01/01 42 3 Within the context of the present invention, the following terms have the following meaning unless expressly specified otherwise: A N-terminal radical of the arginine mimetic, or a radical which is linked N-terminally to the arginine mimetic, is a radical which is bonded to the arginine mimetic by way of the N' atom of the N-terminal amino group of the arginine mimetic or by way of the group in the arginine mimetic which corresponds to the amino group of the unmodified arginine.

Correspondingly, within the context of the present invention, a C-terminal radical of the arginine mimetic, or a radical which is linked C-terminally to the arginine mimetic, is understood as being a radical which is bonded to the arginine mimetic by way of the C-terminal C atom of the carboxyl group of the arginine mimetic or by way of the group in the arginine mimetic which corresponds to the carboxyl group of the unmodified arginine.

Within the context of the present invention, intermolecular interactions are all forms of van der Waals interactions, such as electrostatic interactions between charged radicals in the inhibitor and oppositely charged groups in the factor Xa protein, interactions between polar groups in the inhibitor and oppositely polarized groups in the factor Xa protein, and also hydrophobic interactions between nonpolar groups in the inhibitor and in the factor Xa protein, and the like, and also hydrogen bonds between the inhibitor and the factor Xa protein.

In connection with the present invention, an arginine mimetic is understood as being a compound which possesses the same functional characteristics as arginine or functional characteristics which are similar to those of arginine, e.g. a side chain having .1 WO 01/58859 6 PCT/EP01/014 23 a positive charge at physiological pH, as is characteristic for the guanidinium group of the side chain of arginine. Thus, an arginine mimetic can be an amino acid analog of arginine, i.e. a compound in which the N-terminal amino group, the C-terminal carboxyl group and/or the side chain of arginine has been chemically modified.

Amino acid analogs in which the side chain comprises a substituted or unsubstituted, saturated or unsaturated, carbocylic or heterocyclic radical can, in particular, be used as arginine mimetics in the present invention.

While a ring of this nature is preferably a phenyl ring, it can also be a pyridine ring or a piperidine ring, or another saturated or unsaturated or aromatic, carbocyclic or heterocyclic group, with it being possible for the heteroatom(s) to be nitrogen, oxygen and/or sulfur.

Substituents of such a previously mentioned carbocyclic or heterocyclic radical are preferably basic substituents such as amidino, guanidino, amino, alkylamino, aminoalkyl, amide substituents and the like. It is furthermore also advantageously possible to use polar substituents such as halogens, e.g. chlorine, hydroxyl or alkoxy. The abovementioned carbocyclic or heterocyclic radicals can be substituted once or more than once by the abovementioned substituents, with combinations of the abovementioned substituents also being possible.

Furthermore, within the context of the present invention, the term arginine mimetics also encompasses modifications of the N-terminal amino group and of the C-terminal carbonyl group, with the proviso that these modifications exhibit the same, or essentially the same, spatial configurations as are typical for the unmodified arginine backbone. An example of such a *t 1 WO 01/58859 7 PCT/EP01/01423 modification is the reduction of the C-terminal carbonyl group to a CH 2 group.

In the present invention, it is in principle also possible to use other arginine mimetics which are known from the prior art, or which are derived therefrom, with the proviso that the arginine mimetic meets the steric requirements for a P1 substrate of the factor Xa protein and the C-terminal radical of the arginine mimetic can enter into intermolecular interactions with the S3/S4 pocket of the factor Xa protein.

In the present invention, the greatest preference is given to using, as the arginine mimetic, a phenylalanine analog which is substituted by a basic radical on the aromatic ring. Most preferably, the basic substituent is an amidino group at the 3 position of the aromatic ring.

When an amino acid analog, such as the above-described phenylalanine analog, is used as the arginine mimetic, the above-described, advantageous binding mode of the factor Xa inhibitor according to the invention is achieved by the chirality at the Ca atom of the arginine mimetic, or at a corresponding chiral center of a backbone-modified arginine mimetic, being R, such that the radical which is linked C-terminally to the arginine mimetic extends into the S3/S4 pocket of the factor Xa protein and can there enter into intermolecular interactions with the hydrophobic groups of the S3/S4 pocket.

When used as an arginine mimetic, a (R)-chiral amino acid analog has the additional advantage that it is more stable than (S)-chiral amino acid analogs and that the inhibitor can consequently remain active for a longer period in the body when used pharmacologically.

WO 01/58859 8 PCT/EP01/01423 In a preferred embodiment of the present invention, the C-terminal radical of the arginine mimetic comprises a linker which is bonded directly to the arginine mimetic and also an optionally substituted hydrophobic group which can enter into intermolecular interactions with the hydrophobic S3/S4 pocket of the factor Xa protein.

The linker is preferably of a size which is suitable for bridging the S2 pocket of the factor Xa protein, i.e. its spatial configuration is preferably similar to that of the natural P2 substrate Gly.

The hydrophobic group of the C-terminal radical preferably exhibits a spatial configuration which enables the hydrophobic group to fit optimally into the S3/S4 pocket of the factor Xa protein. In addition, it is advantageous if the hydrophobic group is substituted by one or more basic substituents which are configured such that it is possible for interactions to take place with negatively charged or negatively polarized groups of the factor Xa protein in the neighborhood of the S3/S4 pocket. Preferred basic substituents are amidino, guanidino, amino, alkylamino, aminoalkyl and amide substituents, and the like.

The present invention relates, in particular, to compounds in accordance with the following structural formula I: 0

H

RiT y R2 R3 WO 01/58859 9 PCT/EPOI/01423 in which R comprises a linker which is directly bonded to the phenylalanine analog, -and a substituted or unsubstituted, saturated or unsaturated group R 4

R

2 comprises a linker L 2 which is bonded directly to the phenylalanine analog, and a substituted or unsubstituted, saturated or unsaturated group R 7 and R 3 is a basic substituent at the 3 or 4 position of the aromatic ring of the phenylalanine analog and the aromatic ring is optionally substituted by at least one further substituent R

Y

where z 0 to 4.

The linker L 1 is used for linking the group R 4 to the nitrogen atom of the phenylalanine analog of the formula I. In this connection, L' can be any group which enables such a linkage to take place. Preference is given to L' being a group which is chemically and enzymically stable in order to prevent the compound of the formula I breaking down when being used as a pharmaceutical composition.

The linker L 1 can simply be a bond. In that case,

R

4 is R 4 Preferably, the linker L 1 comprises a group Rx having a chain length of from 1 to 10 atoms, preferably of from 1 to 5 atoms, such as Cl-Cio, in particular C 1 -Cs-alkyl, CI-Cio-, in particular Ci-C 5 alkenyl, Ci-Clo-, in particular C 1 -Cs-alkynyl, with this group also being able to contain heteroatoms, in particular O, S or N, in the chain, e.g. (O-CH 2

-CH

2 )n in which n 1 to 3. Particularly preferably, the linker

L

1 comprises, in addition to said group Rx or without said group Rx, a linking group which is bonded to the nitrogen as phenylalanine analog.

Particularly preferably, R I comprises a linker L 1 which is capable of forming hydrogen bonds. Linkers L' or linking groups which are capable of forming hydrogen bonds, or potential hydrogen acceptors or donors, which are additionally preferred because of their geometry, comprise linkers, such as -CO-NH- or -COO-, WO 01/58859 10 PCT/EP01/01423 which, together with the NH group of the phenylalanine analog, form an amide bond (in the case of a urea bond (in the case of -CO-NH-) or a urethane bond (in the case of An N-terminal linkage of the group R 4 by way of an -S0 2 -linker is likewise possible.

Examples of preferred

R

1 radicals are -CO-R 4

-CO-NH-R

4 or -COOR 4 and corresponding sulfur groups -CS-R 4

-CS-NH-R

4 or -COSR 4 Particularly preferably, R 1 is

=COOR

4 Even more preferably,

R

1

-CO-NH-R

4 The abovementioned group Rx can be arranged between the linking group and the radical R 4 It has been found that urea derivatives

(L

1 -CO-NH-) inhibit FXa outstandingly well and are extremely stable chemically and enzymically, for which reason they are particularly suitable as inhibitors of FXa.

However, the linker L 1 can also be glycine (-CO-CH 2

-NH-)

or another natural or unnatural amine acid

(-CO-CHR-NH-).

The group or radical R 4 is preferably a hydrophobic radical. However, it can also be a hydrophilic radical or a radical which possesses a hydrophobic group which carries one or more hydrophilic substituents.

R

4 can, for example, be a saturated or unsaturated, substituted or unsubstituted, noncyclic alkyl radical; a saturated or unsaturated, substituted or unsubstituted carbocyclic radical; or a saturated or unsaturated, substituted or unsubstituted heterocyclic radical.

R

4 is preferably a Ci-o 3 -alkyl-, C2- 30 -alkenyl-, C 2 30 alkynyl-, C 3 .30-cycloalkyl-, Cs- 30 -aryl-, C3-30heteroaryl-,

C

6

-C

3 o-alkaryl- or Cso 30 -alkheteroaryl radical, with these radicals being able to carry one or more substituents.

R

4 preferably comprises at least 4 C atoms, more preferably at least 6 C atoms and preferably up to 24 C atoms, more preferably up to 18 C atoms. Suitable heteroatoms which the R 4 radical can WO 01/58859 11 PCT/EPOI/01423 contain are, for example, 0, N, S and P. The radical R 4 can furthermore have one or more substituents. R 4 is preferably a noncyclic Ci- to Cs-alkyl radical which is substituted by at least one radical R 6 with R 6 being selected from CnH 2 n+ 1 where n 1 to 10. R 4 is particularly preferably t-butyl. In another preferred embodiment, R 4 is substituted or unsubstituted phenyl, benzyl, fluorenyl, naphthyl, -C(CH 3 2

-C

6 Hs or adamantyl.

Most preferably, R 4 adamantyl.

Particularly preference is furthermore given to R 4 radicals which are larger (on the basis of the volume occupied) than is the phenyl radical.

According to the invention, the linker L 2 which is bonded directly to the phenylalanine analog is preferably of a size which is suitable for bridging the S2 pocket of the factor Xa protein, i.e. its spatial configuration is preferably similar to that of the natural P2 substrate Gly. In this connection, the linker of R 2 is preferably -OR 5

-NH-R

5

-NH-NH-R

5 or

-CH

2

R

5 where R 5 is a substituted or unsubstituted, saturated or unsaturated, carbocyclic, heterocyclic or noncyclic alkyl radical, or can be a group Rx, as defined above. Particularly preferably, the linker is

L

2

-NH-R

s However, the linker L 2 can also simply be a bond.

It is advantageous for a factor Xa inhibitor according to the invention having the structural formula I if a hydrophobic R 5 radical is C-terminally linked to the phenylalanine derivative by way of an ester or amide bond, with R 5 particularly preferably being a substituted or unsubstituted Ci- to Cs-alkyl radical. In this connection, R 5 can have the formula -(CH2)m, in which m 1 to 3. Particularly preferably, R 5 is

-CH

2

-CH

2 Furthermore, R 2 comprises a saturated or unsaturated group R 7 which is unsubstituted or substituted by one or more radicals R 8 and which can be WO 01/58859 12 PCT/EP01/014 23 a noncyclic radical but is, in particular, a carbocyclic radical, such as a cyclic alkyl, alkylaryl, arylalkyl or aryl radical or a heterocyclic radical which contains at least one heteroatom, such as oxygen, nitrogen and/or sulfur, with R 8 preferably being a basic substituent and/or a substituent which functions as a hydrogen bond donor or acceptor, and/or a halogen.

R

7 is preferably a C1- 30 -alkyl, C2_ 30 -alkenyl,

C

2 30 alkynyl, C3- 30 -cycloalkyl, Cso30-aryl,

C

3 30 -heteroaryl,

C

6

-C

30 -alkaryl or C 4 -30-alkheteroaryl radical, with these radicals being able to carry one or more substituents.

R

7 preferably comprises at least 4 C atoms, more preferably at least 6 C atoms and preferably up to 24 C atoms, more preferably up to 18 C atoms. Suitable heteroatoms which the R' radical can contain are, for example, 0, N, S and P. The R 7 radical can additionally possess one or more substituents.

In particular, R can be a phenyl, piperidine, pyrrol, furan, thiophene, pyridine, naphthalene, anthracene or indole radical which is unsubstituted or substituted by one or more R 9 radicals. Other aromatic radicals, including fused aromatic or heteroaromatic radicals, are likewise conceivable.

R

8 is preferably a radical which is a positively charged radical under physiological conditions, e.g. a pH of approx. 6.5-7.5.

Particularly preferably,

R

8 is an amidino, guanidino, amino, ester, alkylamino, aminoalkyl, cyano, amide or hydroxyl radical, or the like.

Particularly advantageously for the use of the compounds according to the invention having the structural formula I as factor Xa inhibitors, a radical -NH-CHR'-COO-(CH2)mR, in which m 1 to 5, R is as defined above and R 9 is a derivatized or nonderivatized WO 01/58859 13 PCT/EP01/01423 side chain of a natural amino acid, and which can readily be produced synthetically -by esterifying a natural or unnatural amino acid, can be used as the radical R 2 In the present invention, R 3 is preferably an amidino, guanidino, amino, alkylamino, aminoalkyl or amide radical, or the like, particularly preferably an amidino radical. The aromatic ring of the phenylalanine radical is substituted at the 3 or/and 4 position, preferably at the 3 position, by the radical R 3 e.g.

an amidino radical.

Furthermore, the ring can also advantageously be additionally substituted by one or more substituents

R

Y

where z 0 to 4. Preference is given to polar substituents, such as halogen, e.g. fluorine, chlorine, bromine, iodine, hydroxyl or alkoxy and/or basic substituents.

RY can preferably, in each case independently at each occurrence, be a halogen, e.g. fluorine, chlorine, bromine, iodine, -OH, -NH 2 -formyl, -acetyl, -OMe (Me methyl), -OEt (Et ethyl), NHMe, -NHEt, SH, SEt, SMe, NMe 2

-CH

3 -CH20H, -CH 2

-CH

3 -NH-OH, -COOH, -COOMe, CN, NO 2 or -CH 2

CH

3 The groups mentioned for the substituent R Y are also preferred substituents for the other substituted groups mentioned herein in the case of R 4

R

5 and Rl), unless explicitly indicated otherwise.

Surprisingly, it has been found that it is advantageous for inhibiting factor Xa with a compound according to the invention of the structural formula if the phenylalanine analog is chiral, because the Cterminal radical is then able to enter into intermolecular interactions with the S3/S4 pocket of the factor Xa protein. Preference is therefore given to PCT/EP01/01 4 2 3 WO 01/58859 14 the abovementioned compounds being in the

(R)

conformation. However, the invention also encompasses the compounds in the conformation, and also mixtures of and enantiomers.

A very powerful inhibitory effect on factor Xa was achieved using the compound according to the invention N-l-adamantyloxycarbonyl-D-3-amidinophenylalanine-(2phenyl)-1-ethylamide.

An even more powerful inhibitory effect on factor Xa was observed in the case of the compound according to the invention N-(l-adamantylaminocarbonyl)-D-3amidinophenylalanine-(2-phenyl)-1-ethylamide.

The compound according to the invention can be present in free form or as a pharmaceutically acceptable salt, for example as a hydrochloride.

In that which follows, the present invention is illustrated using representative compounds according to the invention. Figures 2 to 4 list their inhibitory strengths toward factor Xa and, by comparison, toward uPA, thrombin and trypsin.

The preference of factor Xa for 3-amidinophenylalanine derivatives as compared with 4-amidinophenylalanine derivatives is in agreement with previous publications (Maduskuie et al., see above), although a preference for the 4-amidino group was observed in the case of the inverse binding of the bisbenzamidine compound 2 Gabriel et al., see above). Interestingly, the 4guanidinoderivative 26 in no way meets the steric requirements for a PI radical for this enzyme since a dramatic loss in inhibitory activity is observed as compared with compound 19. This also applies to the other enzymes investigated, i.e. uPA, thrombin and trypsin.

WO 01/58859 15 PCT/EP01/0 1423 Comparison of the racemic compound 15, as a free acid at the C terminus, with the racemic -compound 17, as a C-terminal amide derivative, does not show any important differences, an observation which is in agreement with the crystal structure of des-Gla-factor Xa-complexed DX-9065a, in which the free carboxylate group extends into the surrounding solvent Brandstetter et al. (1996), J. Biol. Chem. 271, 29988-29992). In a similar way, it was observed, in the prior art, that the inhibition of factor Xa is not affected when the compound 2 is present as the ester derivative instead of having a free carboxyl (Gabriel et al., see above). In view of these results, the significantly increased inhibitory effect of compound 11, i.e. the C-terminal methyl ester derivative, as compared with that of compound 15, having the free Cterminal carboxyl group, is extremely surprising.

Evidently, the nature of the bond differs from that of the Daiichi inhibitor DX-9065a (Brandstetter et al., see above) or from that of compound 1 Renatus et al. (1998), J. Med. Chem. 41, 5445 to 5456), in that the ester group is involved in a new type of interaction in the vicinity of the Sl binding site.

The advantageous effect, which is described in the previous paragraph, on the inhibitory action is augmented when, for example, use is made of a Cterminal ester or a C-terminal amide derivative which comprises a hydrophobic group and a linker which is of a suitable size for bridging the S2 pocket of the factor Xa protein, i.e. its spatial configuration is similar to that of the natural P2 substrate Gly. This is illustrated by the compounds according to the invention 29 to 31 (see Figure Thus, in the case of the strongest inhibitor 31, the group -NH-CH2-CH2corresponds to the previously described linker in that it exhibits virtually the same spatial extent as does a Gly radical

-NH-CH

2 The preferred hydrophobic group is a substituted or unsubstituted aryl or wo 01/58859 16 PCT/EPO1/01 4 2 3 alkylaryl group, as previously described, in order to achieve optimal intermolecular interactions with the hydrophobic S3/S4 pocket of the factor Xa protein, and consequently a powerful inhibitory effect.

The effect of the chirality of the 3amidinophenylalanine derivative on the inhibition of factor Xa is depicted using as an example a preferred embodiment of the present invention, i.e. the racemic compound 11. The Ki values of the L-(compound 27) and of the D-enantiomer (compound 28) are listed in Table 3. It is not possible to ascertain any clear preference between the L-(compound 27) and D-enantiomer (compound 28) by carrying out a crystal structure analysis of the known trypsin/11 complex and by carrying out modeling studies for trypsin and factor Xa based on this crystal structure. This is confirmed by the Ki values of trypsin, since trypsin recognizes both enantiomers 27 and 28 with almost identical affinity, and only with a slight preference for the D-enantiomer.

On the other hand, when inhibiting factor Xa, the Denantiomer 28, with a Ki 0.39 uM, surprisingly exhibits an activity which is about 10 times greater than that of the L-enantiomer 27. Whereas the affinity of this type of inhibitor for trypsin and thrombin is only marginally influenced by the chirality of the 3amidinophenylalanine radical, uPA is, on the other hand, only capable of recognizing the L-enantiomer.

This is, therefore, the first report of the inhibitory strength for factor Xa being dependent on the chirality of the arginine mimetic employed and, in particular, the first report of a R-chiral arginine mimetic being an effective inhibitor of factor Xa.

This preference of factor Xa for the chirality of the arginine mimetic is an essential feature of the present invention which it was not possible to expect on the basis of the known investigations. The fact that the chirality at this position has an influence on WO 01/58859 17 PCT/EPO1/01 42 3 selectivity with regard to the inhibition of uPA, trypsin and thrombin, since uPA is .selective for the chirality whereas both trypsin and thrombin recognize both isomers with comparable affinities, is also surprising.

Modeling studies carried out on the complex of factor Xa and compound 11, and based on the known crystal structure of factor Xa Padmanabhan et al. (1993), J. Mol. Biol. 232, 947-966), indicate that the bonding is of the following nature: the benzamidino group of Nl-adamantyloxycarbonyl-D-3-amidinophenylalanine methyl ester (28) lies in the S1 pocket while the adamantyl group is located in a slight recess surrounded by the side chains of Trp 215, Glu 217 and Phe 174 south of the substrate S3/S4 aryl binding site. In this type of binding, the C-terminal ester group points in the direction of the S3/S4 substrate binding pocket, with this being able to explain the preference for the (R) chirality. It can be presumed that a similar binding mechanism also operates in the case of the particularly preferred compounds according to the invention, 29 to 31.

A comparison of hydrophobic head groups which are Nterminally linked to the 3-amidinophenylalanine, such as of the tert-butyl group (compound of the 9fluorenylmethyl group (compound 10), of the 1-adamantyl group (compound 11) and of the benzyl group (compound clearly shows that nonplanar and nonaromatic groups are most suitable. Thus, the compound 11, having the 1adamantyl group, leads to a submicromolar inhibition of factor Xa and, at the same time to remarkable selectivity vis-A-vis uPA, thrombin and trypsin.

While, with the exception of dramatic effects on the inhibition of uPA, the replacement of the N-terminal urethane group, as a potential water bond acceptor in compound 11, with the related urea group (compound 12) PCT/EP01/01 42 3 wo 01/58859 18 evidently does not have any effect on the hydrogen bond network in this segment of the -protease/inhibitor complex, it leads to a desirable stability towards acids, for example stomach acids.

With regard to selectivity, the most marked effects are achieved by means of a free carboxyl group at the C terminus, which group evidently impairs the interactions with uPA, thrombin and trypsin at their active sites in a specific manner. In a similar way, a 4-guanidino group impairs the inhibition not only of factor Xa but also of other trypsin-like enzymes which have been investigated.

The compounds according to the invention are synthesized by means of a process which comprises the following steps: a) adding R 4 -NCO, R 4 -NCS, X-CO-R 4 X-S0 2

-R

4

X-CO-NH-R

4 or X-COOR 4 to D- or L-phenylalanine which possesses the basic substituent

R

3 or a precursor of R 3 at the 3 or 4 position; b) where appropriate converting the precursor of R 3 into the substituent R 3 c) where appropriate adding YR 5 to the reaction product from step b).

In this connection, X can be Cl or an active ester. In the same way, the abovementioned compounds which contain the R 4 radical can be added, if possible, in the form of their respective acid anhydrides.

The N derivatives of the racemic 3- and 4amidinophenylalanine are obtained from the respective 3- and 4-cyano compounds, followed by their conversion into the related amidino derivatives, or by direct derivatization of the amidinophenylalanine. Owing to side reactions which arise as a result of the unprotected amidino group, and owing to difficulties in purifying the hydrophilic amidino compounds, preference PCT/EP01/0 14 2 3 WO 01/58859 19 is given to reaction sequences in which the amidino function is generated in conclusion. -For converting the cyanophenylalanine derivatives into the corresponding amidino compounds, the cyano group can be converted into the amidino radical by adding hydroxylamine hydrochloride and subsequently performing catalytic hydrogenation. However, other modifications of the two step reaction for synthesizing

N-

benzyloxycarbonylamidinophenylalanine piperidide, which have been reported by StOber et al. (Stiber et al.

(1998), Peptide Res. 8, 78-85), are also possible.

When R 2

OR

5 the C derivatives of the racemic 3- and 4-amidinophenylalanine can be obtained by adding the corresponding alcohol, where appropriate in the presence of acid or DCC (dicyclohexylcarbodiimide).

When R 2

NHR

5 it is possible to use the corresponding amine or a corresponding amino acid, where appropriate in the presence of condensing reagents which are customarily used in peptide synthesis. The examples of such condensing reagents are HOBT and TBTU.

An important aspect of the present invention is the use of the compounds according to the invention for producing a composition for anticoagulatory therapy. In connection with the present invention, anticoagulatory therapy is understood as being the treatment of vascular diseases in order to avoid thrombotic vascular occlusions (antithrombotic therapy). Therapies of this nature comprise the prophylaxis and therapy of the venous thromboses and lung embolisms and the antithrombotic therapy of arterial thromboses and embolisms, including coronary heart diseases such as angina pectoris or acute myocardial infarction, cerebrovascular blood flow disturbances, such as transient ischaemic attacks and cerebral infarctions, and peripheral arterial occlusion diseases. In addition, the compounds according to the invention can be used for hemorheologic therapy, i.e. for improving the flowability of the blood.

wo 01/58859 20 PCT/EP01/01 423 The compounds according to the invention of the structural formula I can also be conceived as being suitable inhibitors of other serine proteases, in particular of human thrombin, plasma kallikrein and plasmin. In connection with such an inhibitory effect, the compounds according to the invention can be used for preventing or treating physiological reactions, blood coagulation and inflammatory processes which are catalyzed by the abovementioned class of enzymes.

The present invention furthermore relates to a pharmaceutical composition which, where appropriate, comprises a pharmaceutically acceptable excipient and at least one of the compounds according to the invention. Preference is given to the pharmaceutical composition comprising a therapeutically effective quantity of the compounds according to the invention.

A

"therapeutically effective quantity" is understood as being a quantity of the compounds according to the invention having the structural formula I which, when administered on its own to a mammal, or administered to a mammal in combination with an additional therapeutic agent, exhibits therapeutic activity and is, in particular, active antithrombotically or as an antitumor agent.

Within the context of the present invention, "administration in combination" or "combination therapy" means that the compounds according to the invention of the formula I and one or more additional therapeutic compositions are administered alongside each other to the mammal to be treated. When administration takes place in combination, each component can either be administered at the same time or consecutively at different times in any sequence.

Consequently, each component can be administered separately but sufficiently close to each other chronologically to ensure that they provide the desired Wo 01/58859 21 PCT/EP01/01 42 3 therapeutic effect. Other anticoagulants (or coagulation inhibiting agents) which can be used in combination with the compounds according to the invention include warfarin and heparin and other factor Xa inhibitors which have been described in the prior art.

The administration of the compounds according to the invention in combination with such additional therapeutic compositions can afford an advantage as compared with the respective use of the compounds and compositions on their own by, for example, making it possible to use lower doses in each case, thereby minimizing any possible side effects.

The compounds according to the invention are suitable, in particular, for treatment or prophylactic use in association with diseases which are associated with a pathological expression or overexpression of factor Xa and/or involve an increase in factor Xa proteolytic activity which can in turn be responsible for tumor growth-promoting and metastasis-promoting fibrin depositions.

Thus, the compounds according to the invention are able to efficiently inhibit and/or prevent the growth and/or spread of malignant tumors and the metastasis of tumors. The invention therefore also relates to the use of the compounds according to the invention for producing an antitunor agent. In this connection, the factor X. inhibitors according to the invention can, where appropriate, be formulated together with suitable pharmaceutical auxiliary substances or carrier substances for the purpose of producing drugs. It is furthermore possible, where appropriate, to use the factor Xa inhibitors together with other tumor agents or other active compounds or with other types of treatment, for example in combination with irradiation or surgical interventions. Tumors which exhibit factor WO 01/58859 22 PCT/EP01/01 4 2 3 Xa activity, and which are suitable for being treated with the compounds according to the invention, are, in particular, lung, bladder, liver and ovarian carcinomas, and also malignant melanomas and neuroblastomas.

The compounds according to the invention already inhibit FXa at low concentrations. For example, the compound 31 which is presented herein inhibits with an inhibitor constant Ki 0.074 pM. Whereas the desired FXa inhibition already takes place at such low concentrations, blood coagulation (according to the APPT test) is only affected at substantially higher concentrations of the compounds according to the invention. As a result, the compounds according to the invention can be used selectively for inhibiting FXa without blood coagulation being affected at the same time. In this way, it is possible to use the compounds according to the invention for controlling cancer (which control is connected with the inhibition of FXa) while being able to avoid side-effects, such as bleeding (which is connected to blood coagulation).

This constitutes a fundamental advantage of the compounds according to the invention as compared with other FXa inhibitors, such as the known DX-9065a (Kakkar et al., J. Clinical Pathology Clinical Molecular Pathology Edition 48(5):M288-M290, 1995; Gouinthibault et al., British Journal of Haematology 90(3): 669-680; Nakata et al., Cancer Letters 122(1-2): 127-133, 1998; Yoshida et al., Fibrinolysis Proteolysis, 11(3): 147-154, 1997; Barendszjanson et al., Tumor Biology, 19(2): 104-112, 1998; Donnelly et al., Thrombosis Haemostasis, 79(5): 1041-1047, 1998; Fielding et al., Blood, 91(5): 1802-1809, 1998; Tanabe et al., Thrombosis Research, 96(2): 135-143, 1999).

The pharmaceutical composition can be administered to human and animals in all known ways, for example topically, orally, rectally or parenterally, for wo 01/58859 23 PCT/EP01/01 4 23 example subcutaneously or intravenously. In addition, it can also be administered in the- form of tablets, sugar-coated tablets, capsules, pellets, suppositories, solutions or transdermal systems, such as plasters.

The compounds according to the invention can also be used as standard or reference compounds, for example as a quality standard or control in tests or assays which include the inhibition of factor These compounds can be provided in a commercial kit, for example for use in pharmaceutical research encompassing factor Xa.

The compounds according to the invention can also be used in diagnostic assays which include factor Xa.

The compounds according to the invention can be administered in oral dosage forms such as tablets, capsules, pills, powders, granules, elixirs, tinctures, suspensions, syrups and emulsions. They can also be administered intravenously, intraperitoneally, subcutaneously or intramuscularly, in each case using dosage forms which are well known to the skilled person. While they can be administered on their own, preference is given to administering them together with a pharmaceutical excipient which is selected on the basis of the chosen route of administration and customary pharmaceutical procedures.

The dose of the compounds according to the invention will naturally depend on a variety of known factors, such as the pharmacodynamic characteristics of the particular composition and its nature and the route of administration; and on the species, the age, the sex, the health, the medical condition and the weight of the recipient, and other known factors. A skilled person is able, without further instruction, to determine the quantity of the compound according to the invention which is effective for producing a composition for antithrombotic therapy.

wo 01/58859 24 PCT/EPO1/01 42 3 In general, when being used *to achieve the abovementioned effects, the daily oral dose of the respective active constituents will be in the range of about 0.001 to 1 000 mg/kg of body weight, preferably of from about 0.01 to 100 mg/kg of body weight, per day and, most preferably, from about 1.0 to 20 mg/kg per day. For intravenous administration, the doses which are most preferred are in a range from about 1 to about 10 mg/kg/min during an infusion at a constant rate. The compounds according to the invention can be administered in a single daily dose or subdivided into doses which are given 2, 3 or 4 times daily.

The compounds according to the invention can also be administered in intranasal form or administered transdermally.

The compounds according to the invention are typically suitably selected in admixture with suitable pharmaceutical diluents, excipients or vehicles (which are jointly termed pharmaceutical vehicles in that which follows), with regard to the intended form of administration and in agreement with conventional pharmaceutical procedures.

Examples, in the case of oral administration in the form of a tablet or capsule, the active compound component, in the form of the compound according to the invention, can be combined with an oral, nontoxic, pharmaceutically acceptable, inert vehicle such as lactose, starch, sucrose, glucose, methyl cellulose, magnesium stearate, dicalcium phosphate, calcium sulfate, mannitol, sorbitol and the like. For oral administration in liquid form, oral active compound components can be combined with any oral, nontoxic, pharmaceutically acceptable inert vehicles such as ethanol, glycerol, water and the like.

wo 01/58859 25 PCT/EP01/01423 Furthermore, if necessary or desired, it is also possible to use suitable binders, lubricants, disintegrants and dyes in the pharmaceutical composition. Suitable binders include starch, gelatin, natural sugars such as glucose or beta-lactose, natural and synthetic rubbers, carboxymethyl cellulose, polyethylene glycol, waxes and the like. Lubricants which are used in these dose forms include sodium oleate, sodium stearate, magnesium stearate, sodium benzoate, sodium acetate, sodium chloride and the like.

The disintegrants comprise, inter alia, starch, methyl cellulose, agar, bentonite and the like.

The compounds according to the invention can also be administered in the form of liposomal transport systems. Liposomes can be formed from a large number of phospholipids, such as cholesterol, stearylamine or phosphatidylcholins.

The compounds according to the invention can also be linked to soluble polymers acting as active compound vehicles. These polymers include polyvinylpyrrolidone, pyran copolymer, polyhydroxypropylmethacrylamide phenol, polyhydroxyethylaspartamide phenol or polyethylene oxide polylysine which is substituted by palmitoyl radicals. Furthermore, the compounds according to the invention can be coupled to a number of biodegradable polymers which are useful for achieving controlled release of an active compound, for example to polyglycolic acid, polylactic acid, copolymers of polyglycolic acid and polylactic acid, polyepsiloncaprolactone, polyhydroxybutyric acid, polyorthoesters, polyacetals and the like.

Lipid dosage forms for oral administration can comprise dyes or flavorings for the purpose of increasing patient acceptance.

I

WO 01/58859 26 PCT/EP01/01 4 2 3 In general, water, a suitable oil, salt solutions, aqueous dextrose (glucose) and related sugar solutions, and glycols, such as propylene glycol or polyethylene glycols, are suitable vehicles for parenteral solutions. Solutions for parenteral administration preferably contain a water-soluble salt of the active constituent, suitable stabilizers and, if necessary, buffering substances. Antioxidants, such as sodium disulfite, sodium sulfite or ascorbic acid, either alone or in combination, are suitable stabilizers.

Other suitable pharmaceutical vehicles are described in "Remington's Pharmaceutical Sciences", Mack Publishing Company, which is a standard reference work in this area.

The compounds according to the invention can also be employed as lead substances. They can, in particular, be used for developing or finding other effective factor Xa inhibitors, for example using appropriate algorithms, which may, where appropriate, be computerassisted. When employed as lead substances, the compounds according to the invention can be used, in particular, for developing novel antithrombotic and antitumor agents.

The following examples serve to explain the invention without restricting it in any way.

Examples All the solvents and reagents which are used in the following examples were of the highest commercially available quality and, if required, were further purified and dried using standard methods. Analytical HPLC was carried out on ET 125/4 Nucleosil 100/Cs columns (Macherey-Nagel, DUren, Germany) using a linear gradient of MeCN/2% H 3

PO

4 of 5:95 to 80:20 in 12 minutes. ESI-MS spectra were recorded on a Perkin Elmer Wo 01/58859 27 PCT/EP01/01423 API 165 mass spectrometer (Perkin Elmer, Langen, Germany) .TLC was carried out on silica gel 60 plates using the following solvent systems: CHC1 3 /MeOH/AcOH, 40:10:2; CHCl 3 /MeOH/AcOH, 20:20:1; AcOEt/n-BuOH/H 2 0/AcOH, 10:6:2:2; CHC1 3 /MeOH/AcOH, 190:10:2; CHCl 3 /MeOH/NH3, 20:20:9; CHCl 3 /MeOH/AcOH, 10:20:1; n-hexane/AcOEt/AcOH, 49:49:2.

The synthesis, and inhibitor constants, of the compounds 1 and 4 are taken from the prior art Gabriel (1998), Doctoral Thesis, Technische Universitat Monchen [Munich Technical University]).

N, N' .Dibenzyloxycarbonyl-N"-trifylguaflidine was synthesized in accordance with Feichtinger et al. (J.

Org. Chem. 63, 3804-3805, 1998) 4-Nitrophenylalanine was obtained from Bachem (Heidelberg, Germany) D,L-3- Cyanophenylalanine and D, L-4-cyanophenylalanine were obtained from Sennchemicals (Oielsdorf, Switzerland), while Boc-D-3-cyanophenylalanine and Boc-L-3cyanophenylalanine were obtained from Syntetech (Albany, Oregon, USA) The two latter compounds were N4-deprotected in 95% TFA (trifluoroacetic acid).

PCT/EPO1/01 4 2 3 WO 01/58859 28 Example 1: Synthesis of N-tert-butyloxycarbonyl-D,L-3cyanophenylalanine (Boc) 2 0 (5.74 g; 26.29 mmol) in dioxane (5 ml) was added to a stirred solution of D,L-(3cyano)phenylalanine (5 g; 26.29 mmol) in dioxane ml) and 1 M NaOH (26.3 ml). After one hour, the solution was evaporated and the residue was partitioned between AcOEt and 5% aqueous

KHSO

4 solution. The aqueous phase was extracted three times with AcOEt and the combined organic phases were dried (over Na 2 SO4) and evaporated, resulting in a pale yellow oil, which crystallized at 4 0

C.

Yield: 6.9 g TLC (solvent system Rf 0.77; HPLC: tR 8.2 min; MS m/z 291.0 calculated Mr 290.1.

Example 2: Synthesis of N-tert-butyloxycarbonyl-D,L- 3 hydroxyamidinophenylalanine (6) A solution of compound 5 (1 g; 3.44 mmol), hydroxylamine hydrochloride (359 mg; 5.17 mmol) and KOH (483 mg; 8.6 mmol) in EtOH (50 ml) was boiled under reflux overnight. Insoluble KCl was filtered off and the solution was evaporated and the residue was dissolved in water (30 ml) and acidified to pH 2.5 with 1 M HC1. The solution was washed twice with AcOEt ml) and the product was subsequently extracted five times with water-saturated n-BuOH. The combined n-BuOH layers were evaporated.

Yield: 870 mg of white foam; TLC (solvent system Rf 0.62; HPLC: tR 5.3 min; MS m/z 324.0 calculated Mr 323.2.

Example 3: Synthesis of N-tert-butyloxycarbonyl-D,L-3amidinophenylalanine hydrochloride (7) wo 01/58859 29 PCT/EP01/014 2 3 The compound 6 (870 mg; 2.69 mmol) was hydrogenated in water (50 ml) over 10% Pd/C at 50 0 C for a period of h. The catalyst was filtered off and the solution was evaporated to dryness in the added presence of 1 M HC1 (2.7 ml).

Yield: 710 mg TLC (solvent system Rf 0.18; HPLC: ta 5.4 min; MS m/z 308.4 calculated Mr 307.2.

Example 4: Synthesis of N-tert-butyloxycarbonyl-D,L-3amidinophenylalanine methyl ester hydrochloride (8) A solution of 7 in MeOH (5 ml) was acidified down to pH 2 with 6 M HC1 and stirred at room temperature for 24 h. The solution was evaporated down to dryness.

Yield: quantitative; TLC (solvent system Rf 0.41; HPLC: t, 5.9 min; MS m/z 322.4 calculated Mr 321.2.

Example 5: Synthesis of D,L-3-amidinophenylalanine methyl ester dihydrochloride (9) A solution of 8 (230 mg; 0.64 mmol) in 6 M HC1 in dioxane (5 ml) was stirred at room temperature. After 1 h, the solution was evaporated down to dryness.

Yield: quantitative; TLC (solvent system Rf 0.10; MS m/z 222.2 calculated Mr 221.1.

Example 6: Synthesis of N-9-fluorenylmethyloxycarbonyl- D,L-3-amidinophenylalanine methyl ester hydrochloride Fmoc-Cl (9-fluorenylmethoxycarbonyl chloride, 44 mg; 0.17 mmol) and TEA (triethylamine, 24 l; 0.17 mmol) were added to a solution of compound 9 (50 mg; 0.17 mmol) in DMF (500 After 30 min at room WO 01/58859 30 PCT/EPO1/014 2 3 temperature, TEA (12 p1) was added in order to complete the reaction. After 3 h, the solvent -was evaporated and the residue was dissolved in water. After acidifying down to pH 3 with 1 M HC1, the product was collected by centrifugation and precipitated once again from AcOEt/diisopropyl ether.

Yield: 60 mg TLC (solvent system Rf 0.58; HPLC: tR 8.0 min; MS: m/z 444.0 calculated Mr 443.2.

Example 7: Synthesis of N-1-adamantyloxycarbonyl-D,L-3amidinophenylalanine methyl ester hydrochloride (11) Compound 11 was prepared essentially as described for compound 10 using Adoc-F (1-adamantyloxycarbonyl fluoride) and was reprecipitated from AcOEt/diisopropyl ether.

Yield: 86%; TLC (solvent system Rf 0.55; HPLC: tR 7.6 min; MS: m/z 400.4 calculated Mr 399.2.

Example 8: Synthesis of N-1-adamantylaminocarbonyl-D,L- 3-amidinophenylalanine methyl ester hydrochloride (12) Compound 9 (46 mg, 0.156 mmol) was reacted for 3 h in DMF (500 p1) containing 1-adamantyl isocyanate (27.7 mg; 0.156 mmol) and TEA (22 pl, 0.156 mmol).

After the solvent had been evaporated, the residue was crystallized from isopropanol/diisopropyl ether.

Yield: 55 mg HPLC: tR 8.7 min; MS m/z 399.4 (M+H) calculated Mr 398.2.

Example 9: Synthesis of N-1-adamantyloxycarbonyl-D,L-3cyanophenylalanine (13) A solution of D,L-(3-cyano)phenylalanine (2 g; 10.5 mmol), Adoc-F (2.08 g; 10.5 mmol) and 2 M NaOH WO 01/58859 31 PCT/EPO1/O1423 (7.8 ml; 15.6 mmol) in dioxane (50 ml) was stirred at room temperature for 3 h. The residue was partitioned between AcOEt and 5% aqueous

KHSO

4 solution. The aqueous phase was extracted three times with AcOEt and the combined organic phases were washed with salt solution, dried (over Na 2

SO

4 and evaporated. The resulting yellowish oil was treated with diethyl ether and evaporated down to a white foam.

Yield: 3.7 g HPLC: tR 8.7 min; TLC (solvent system Rf 0.72; MS m/z 369.5 calculated Mr 368.2.

Example 10: Synthesis of N-1-adamantyloxycarbonyl-D,L- 3-hydroxyamidinophenylalanine hydrochloride (14) Compound 13 (3.7 g; 10 mmol) was reacted with hydroxylamine hydrochloride and worked up as described for compound 6.

Yield: 3.9 g HPLC: tR 9.1 min; TLC (solvent system Rf 0.66; MS m/z 402.4 calculated Mr 401.2.

Example 11: Synthesis of N-l-adamantyloxylcarbonyl-D,L- 3-amidinophenylalanine hydrochloride The catalytic reduction of compound 13 (3.9 g; 9.7 mmol) was carried out as described for compound 7.

Yield: 3.5 g HPLC: tR 9.4 min; MS: m/z 386.4 calculated Mr 385.2.

Example 12: Synthesis of N-l-adamantyloxycarbonyl-D,L- 3-cyanophenylalanine piperidide (16) SOC1 2 (120 pl; 1.63 mmol) was added dropwise, at 0 C and while stirring vigorously, to a solution of compound 13 (300 mg; 0.814 mmol) and piperidine wo 01/58859 32 PCT/EPO1/01 42 3 (480 pl; 4.88 mmol) in methylene chloride (5 ml). After the mixture had been allowed to -warm up to room temperature, and after 2 h, the reaction mixture was diluted with methylene chloride and washed with aqueous NaHCO 3 5% aqueous

KHSO

4 solution, water and salt solution, and dried (over Na 2

SO

4 The solution was brought to dryness.

Yield: 240 mg TLC (solvent system Rf 0.76; MS m/z 436.2 calculated Mr 435.3.

Example 13: Synthesis of N-l-adamantyloxycarbonyl-D,L- 3-amidinophenylalanine piperidide hydrochloride (17) The reaction of compound 16 (240 mg; 0.55 mmol) with hydroxylamine hydrochloride, and the following catalytic reduction to give the compound 17, were carried out as described for compounds 6 and 7.

Yield: 84 mg (31% over the two steps); HPLC: tR 10.0 min; MS: m/z 453.4 calculated Mr 452.3.

Example 14: Synthesis of N-tert-butyloxycarbonyl-D,L- (4-amidino)phenylalanine hydrochloride (18) Compound 18 was synthesized starting from D,L-(4cyano)phenylalanine, as described for the 3-substituted phenylalanine 7.

Yield: 57% (over 3 steps); HPLC: tR 5.2 min; MS m/z 308.4 calculated Mr 307.2.

Example 15: Synthesis of D,L-4-amidinophenylalanine dihydrochloride (19) The deprotection of compound 18 (625 mg; 2.037 mmol) was carried out in 6 M HC1 in dioxane as described for compound 9.

WO 01/58859 33 PCT/EP01/0l 42 3 Yield: 540 mg TLC (solvent system Rf 0.23; MS: m/z 208.3 calculated Mr 207.2.

Example 16: Synthesis of D,L-4-amidinophenylalanine methyl ester dihydrochloride SOC12 (180 l; 2.47 mmol) was added dropwise, at and while stirring vigorously, to a solution of compound 17 (230 mg; 0.824 mmol) in MeOH (2 ml). The reaction mixture was allowed to warm to room temperature and was stirred for 18 h. The solvent was evaporated and the product was crystallized from EtOH/diethyl ether.

Yield: 182 mg TLC (solvent system Rf 0.54; MS: m/z 222.4 calculated Mr 221.1.

Example 17: Synthesis of N-l-adamantyloxycarbonyl-D,L- 4-amidinophenylalanine methyl ester hydrochloride (21) Compound 21 was prepared from compound 20 using Adoc-F as described for compound Yield: 32 mg HPLC: tR 9.4 min; MS: m/z 400.4 calculated Mr 399.2.

Example 18: Synthesis of D,L-4-nitrophenylalanine methyl ester hydrochloride (22) SOC1 2 (1.27 il; 18.88 mmol) was added dropwise to an ice-cool solution of (4-nitro)phenylalanine (1 g; 4.72 mmol) in MeOH (5 ml). After 20 h, the solvent was evaporated and the weakly yellow solid was washed with ether and dried.

Yield: 1.19 g TLC (solvent system Rf 0.70; MS: m/z 225.2 calculated Mr 224.1.

I

wo 01/58859 34 PCT/EPO1/0l 42 3 Example 19: Synthesis of N-l-adamantyloxycarbonyl-D,L- 4-nitrophenylalanine methyl ester (23) Compound 23 was prepared from compound 22 using Adoc-F as described for compound Yield: 725 mg HPLC: tR 12.7 min; MS: m/z 403.4 calculated Mr 402.2.

Example 20: Synthesis of N-1-adamantyloxylcarbonyl-D,L- 4-aminophenylalanine methyl ester (24) Compound 23 (725 mg; 1.8 mmol) was hydrogenated for 2 h over Pd/C in MeOH (20 ml); the catalyst was subsequently filtered off and the solvent was evaporated. The crude product was chromatographed through silica gel (eluent: n-hexane/AcOEt/AcOH, 49:49:2).

Yield: 556 mg HPLC: tR 14.2 min; TLC (solvent system Rf 0.54; MS: m/z 373.4 calculated Mr 372.4.

Example 21: Synthesis of N-l-adamantyloxycarbonyl-D,L- 4- N-dibenzyloxycarbonyl)guanidinophenylalanine methyl ester A solution of compound 24 (57 mg; 0.153 mmol), N,N'dibenzyloxycarbonyl-N"-trifylguanidine (70 mg; 0.153 mmol) and TEA (21 al; 0.153 mmol) in methylene chloride (500 pL1) was stirred at 50 0 C for three days in a sealed reaction vessel equipped with a screw closure.

The solvent was evaporated and the crude product in AcOEt (10 ml) was washed twice with 5% aqueous KHSO 4 water and salt solution. The organic phase was dried (over Na 2

SO

4 and evaporated.

Yield: 96 mg of a colorless oil; HPLC: tR 14.5 min; MS: m/z 683.4 calculated Mr 682.3.

I

WO 01/58859 35 PCT/EP01/0 1423 Example 22: Synthesis of N-l-adamanty1oxycarbolyl-D,L- 4-guanidinophenylalanile methyl ester hydrochloride (26) Compound 26 was obtained by catalytically hydrogenating compound 25 (96 mg; 0.141 mmol) over Pd/C in MeCH ml) which contained 1 Mi HCl (140 iii; 0. 141 mmol) The catalyst was filtered off, the solution was evaporated and the residue was recrystallized from AcOEt/diisopropyl ether.

Yield: 53 mg HPLC: tR 9.5 min; MS m/z 415.4 calculated M, 414.2.

Example 23: Synthesis of N-1-adamantyloxycarbony1-L- 3 amidinophenylalaiine methyl ester hydrochloride (27) Compound 27 was synthesized as described for compound 11, preceding from L- (3-cyano)phenylalanile.

HPLC: tR 9.6 min; MS: m/z 400.2 calculated M, 399.2.

Example 24: Synthesis of N- 1-adamant yloxycarbonyl-D- 3 amidinophenylalanine methyl ester hydrochloride (28) Compound 28 was synthesized as described for compound 11, proceeding from D-(3-cyano)pherlylalanine.

HPLC: tR 8.2 min; MS: m/lz 400.4 calculated Mr 399.2.

Example 25: Synthesis of N-1-adamantyloxycarbonylD- 3 amidinophenylalaline propylamide hydrochloride (29) Na-Adoc-D-3-Amidinophenylalanine hydrochloride (50 mg; 0.-118 mmol) n-propylamine (29 ptl; 0. 354 mmcl) and HOBT (19 mg; 0. 142 mmol) were dissolved in 2 ml of DMF. TBTU (46 mg; 0.142 mmol) was added and the reaction mixture WO 01/58859 36 PCT/EP01/01423 was stirred at room temperature for 3 h. After the solvent had been evaporated off in vacuo, the resulting oil was dissolved in 20 ml of ethyl acetate. The product began to precipitate out immediately and was separated off by centrifugation. The colorless solid was washed with ethyl acetate and dried in vacuo.

Yield: 26 mg HPLC: tR 7.8 min; MS: m/z 427.4 calculated Mr 426.3.

Example 26: Synthesis of N-l-adamantyloxycarbonyl-D- 3 amidinophenylalanine benzylamide hydrochloride Na-Adoc-D-3-Amidinophenylalanine hydrochloride (30 mg; 0.071 mmol), benzylamine (23 pl; 0.213 mmol) and HOBT (11 mg; 0.085 mmol) were dissolved in 2 ml of DMF. TBTU (27 mg; 0.085 mmol) was added and the reaction mixture was stirred at room temperature. After 3 h, the precipitated salt was filtered off and the solvent was evaporated in vacuo. The residue was dissolved in 10 ml of ethyl acetate, after which this solution was washed with 5% aqueous NaHCO 3 solution and salt solution and dried over anhydrous Na 2

SO

4 After the solvent had been evaporated off in vacuo, the crude product was dissolved in 1 ml of ethyl acetate. 10 pl of 6N HC1 in dioxane were added and the product was precipitated with tert-butyl methyl ether. The flocculent precipitate was washed with diethyl ether and dried in vacuo.

Yield: 16 mg HPLC: tR 10.3 min; MS: m/z 475.2 calculated Mr 474.3.

Example 27: Synthesis of N-l-adamantyloxycarbonyl-D-3amidinophenylalanine-(2-phenyl)-1-ethylamine hydrochloride (31) Na-Adoc-D-3-Amidinophenylalanine hydrochloride (30 mg; 0.071 mmol), phenethylamine (27 pl; 0.213 mmol) and WO 01/58859 37 PCT/EP01/014 2 3 HOBT (11 mg; 0.085 mmol) were dissolved in 2 ml of DMF.

TBTU (27 mg; 0.085 mmol) was added and the reaction mixture was stirred at room temperature. After 3 h, the reaction had not come to an end and 15 mg of TBTU (0.047 mmol) were added and the mixture was stirred for a further 3 h. The solvent was evaporated in vacuo. The residue was dissolved in 10 ml of ethyl acetate and this solution was washed three times with 5% aqueous NaHCO 3 solution and 1 x with 2 ml of 0.5 N HC1 and dried over anhydrous Na 2

SO

4 After the solvent had been evaporated off in vacuo, the product was precipitated from iPrOH/DIPE. The flocculent precipitate was washed with diethyl ether and dried in vacuo.

Yield: 10 ml HPLC: tR 6.8 min; MS m/z 489.4 calculated Mr 488.3.

Example 27a: Compounds 32 to 36 were synthesized in an analogous manner to the above-described preparation methods.

Compound 32, containing the linker -NH-CO-NH-, inhibits FXa better, by a factor of 3, than does compound 31, containing the linker -0-CO-NH-.

Example 28: Determining the inhibitor constants The measurements were carried out at 25 0 C on a microplate reader (MR 5000, Dynatech, Denkendorf, Germany). The test medium consisted of 200 p1 of Tris buffer (0.05 M; 0.154 M NaCl, 5% ethanol, pH .1 of aqueous substrate solution and 50 pl of enzyme solution. Two concentrations of the substrate and five concentrations of the inhibitor were used. Three minutes after adding the enzyme, 25 1p of acetic acid were added in order to quench the reaction and the optical density was measured at 405 nm. The Ki values were calculated in accordance with Dixon Dixon (1953), Biochem. J. 55, 170-171) using a WO 01/58859 38 PCT/EPOI/01423 linear regression. The Ki values given in Figures 2 and 3 are means of at least three determinations.

Example 29: Enzymes and substrates for the Ki determination The following enzymes and the corresponding substrates were used at the given final concentrations: bovine thrombin, prepared in accordance with Walsmann Walsmann (1968), Pharmazie 23, 401-402) (2 262 U/mg, final concentration 0.45 U/ml), substrate MeSO 2 -D-hexahydrotyrosyl-Gly-Arg-pNA (final concentration 0.18 and 0.09 mM); bovine factor Xa (5 U/vial, 0.11 U/ml; Diagnostic Reagents Ltd., Thame, UK), substrate MeSO 2 -D-Nle-Gly-Arg-pNA (0.36 and 0.18 mM); human factor Xa (0.18 g/ml; Kordia Lab. Supplies, Leiden, Netherlands), substrate as for bovine factor Xa human plasmin (0.67 CTA U/mg, 0.06 CTA U/ml; Behringwerke AG, Marburg, Germany), substrate Tos-Gly- Pro-Lys-pNA (0.18 and 0.09 mM); human uPA (500 000 U/vial, final concentration 150 U/ml; Ribosepharm GmbH Haan, Germany), substrate Bz-pAla-Gly-Arg-pNA (0.18 and 0.09 mM); bovine pancreas trypsin (42 U/mg, 0.0038 U/ml; Serva, Heidelberg, Germany), substrate MeSO 2

-D-

hexahydrotyrosyl-Gly-Arg-pNA (0.18 and 0.06 mM).

The substrates were supplied by Pentapharm Ltd., Basel, Switzerland.

Figures Figure 1: Structures and inhibitor strengths of factor Xa inhibitors 1, 2 and 3 from the prior art.

P \OPERMl005\2560164 -nd rpc doc-29A -39- Figure 2: Derivatives of 3-/4-amidino- or 4guanidinophenylalanine, respectively, containing N a substituted carbamate or urea.

Figure 3: Enantiomerically pure derivatives of 1adamantyloxycarbonyl-3-amidinophenylalanine methyl ester. Ki values marked with correspond to Ki values for human factor Xa.

Figure 4: Derivatives of N-l-adamantyloxycarbonyl-3amidinophenylalanine containing Ca-substituted amide. Ki values marked with correspond to Ki values for human factor Xa.

Figure 4a: Other derivatives of N-1adamantyloxycarbonylamidinophenylalanine containing C a substituted amide, with compounds containing different R 2 radicals, and their Ki values, being depicted.

20 Figure 5: N-(1-Adamantylaminocarbonyl)-D-3-amidinophenylalanine-(2-phenyl)-1-ethylamide and its Ki values.

The reference to any prior art in this specification is not, and should not be taken as, an acknowledgment or any form of suggestion that that prior art forms part of the common o* general knowledge in Australia.

Throughout this specification and the claims which follow, 0 unless the context requires otherwise, the word "comprise", 30 and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

Claims (19)

1. A compound in accordance with the following structural formula I: H O N R N R2 Rz Y R 3 I in which: R 1 comprises a linker L 1 which is directly bonded to the phenylalanine analog and a group R 4 wherein L 1 is selected from a bond, a group Rx having a chain length of from 1 to 10 atoms, -CO-NH-, -COO-, -CS-NH-, -COS-, -CO-CH 2 -NH- or a natural or unnatural amino acid; and wherein R 4 is a substituted or unsubstituted, saturated or unsaturated, non- cyclic alkyl radical, carbocyclic radical or heterocyclic radical; 2 R 2 comprises a linker L 2 which is bonded directly to the phenylalanine analog and a group R wherein L 2 is selected from -OR 5 -NH-R 5 -NH-NH-R 5 or -CH 2 -R 5 where R 5 can be a substituted or unsubstituted, saturated or unsaturated, carbocyclic, heterocyclic or noncyclic alkyl radical or a group Rx having a chain length of from 1 to 10 atoms; and wherein R 7 is a saturated or unsaturated carbocyclic radical or heterocyclic radical containing at least one heteroatom, which is unsubstituted or substituted by one or more R 8 radicals, and R 8 is a basic substituent and/or a substituent functioning as a hydrogen bond donor or acceptor and/or a halogen; and R 3 is a basic substituent at the 3 position of the aromatic ring of the phenylalanine radical and the aromatic ring is optionally substituted by additional substituents R Y which are P AOPERMaflD5'm6016 .md -41- selected from polar and/or basic substituents and z being 0 to 4.

2. A compound according to claim 1, in which the linker L 1 is capable of forming hydrogen bonds.

3. A compound according to any one of claim 1 or 2, in which the linker L' is -CO-NH- or -COO-.

4. A compound according to any one of claims 1 to 3, in which R 4 is a non-cyclic Ci- to C5-alkyl radical which is substituted by at least one R 6 radical and R 6 is selected from CnH 2 n+ 1 in which n 1 to A compound according to claim 4, in which R 4 is t- butyl.

6. A compound according to any one of claims 1 to 3, in which R 4 is phenyl, benzyl, fluorenyl or adamantyl.

7. A compound according to any one of claims 1 to 6, in 5 which R 5 has the formula -(CH 2 in which m 1 to 3.

8. A compound according to any one of claims 1 to 7, in which R 7 is a phenyl, piperidine, pyrrol, furan, thiophene, a. pyridine, naphthalene, anthracene or indole radical which is unsubstituted or substituted by one or more R 8 radicals. o 8

9. A compound according to claim 1, in which R 8 is an amidino, guanidino, amino, alkylamino, aminoalkyl or amide radical. radical. PAOPER\M.D20SUS60164 md sp doc.29M6dO5 -42- A compound according to any one of claims 1 to 9, in which R 2 is a -NH-CHR9-COO-(CH 2 )mR 7 radical in which m 1 to R 7 is defined as in claim 9 and R 9 is a derivatized or underivatized side chain of a natural amino acid.

11. A compound according to any one of claims 1 to 10, in which R 3 is an amidino, guanidino, amino, alkylamino, aminoalkyl or amide radical.

12. A compound according to any one of claims 1 to 11, in which the phenylalanine radical is (R)-chiral.

13. A compound according to any one of claims 1 to 12, which is selected from N-(l-Adamantylaminocarbonyl)-D-3- amidinophenyl-alanine-(2-phenyl)-1-ethylamide, N-1- Adamantyloxycarbonyl-D-3-amidinophenylalanine-(2-phenyl)-1- ethylamide, N-1-Adamantyloxycarbonyl-D-3-amidinophenylalanine propylamide, or N-l-Adamantyloxycarbonyl-D-3- amidinophenylalanine benzylamide. S

14. A compound according to any one of claims 1 to 13, as b a pharmaceutically acceptable salt.

15. A compound according to claim 14, wherein the pharmaceutically acceptable salt is the hydrochloride salt. S a

16. A process for preparing a compound according to any one of claims 1 to 13, comprising the following steps: a) adding R 4 -NCO, R 4 -NCS, X-CO-R 4 X-S0 2 -R 4 X-CO-NH-R 4 or X-COOR to D- or L-phenylalanine which, at the 3 or 4 position, possesses the basic substituent R 3 or a precursor of R 3 PAOPERVMAIUOOUW 164 md spcd Ic- 37M5 -43- b) optionally converting the precursor of R 3 into the substituent R 3 and c) optionally adding YR 5 to the reaction product of step b); wherein X is Cl or an active ester, Y is OH when R 2 5 2 5 2 OR 5 Y is H 2 N when R NHR or Y H 2 NNH 2 when R 2 H 2 NNHR 5 and where R 3 R 4 and R 5 are as defined in claim 1.

17. The process according to claim 16, in which X-CO-R 4 X-S0 2 -R 4 X-CO-NH-R 4 or X-COOR 4 is added in the form of its respective acid anhydride.

18. The use of a compound according to any one of claims 1 to 15, for producing an antitumor composition or as a diagnostic agent.

19. Use of a compound according to any one of claims 1 to 15, in the manufacture of a medicament for treating and/or preventing the growth or spread of tumors.

20. A method for treating and/or preventing the growth or spread of tumors, including the step of administering a *ee compound according to any one of claims 1 to

21. A compound according to any one of claims 1 to substantially as hereinbefore described. DATED this 13th day of July, 2005 Wilex AG By DAVIES COLLISON CAVE Patent Attorneys for the Applicants